Google is reportedly developing a more efficient chip for Gemini

Alphabet, Google’s parent company, is designing a new server chip meant to help its in-house Gemini models run more efficiently. The chip, internally dubbed “Frozen v2,” is slated to be released sometime in 2028, The Information reported, citing anonymous sources. According to that report, the chip could be between six and 10 times more efficient than Google’s existing AI chips, measured by the number of tokens generated per unit of power.

Asked about the report, Google did not directly confirm it in a statement to TechCrunch — nor did the company deny it. Instead, it pointed to its broader approach to hardware development: “Our teams are constantly researching and experimenting with new innovations to deliver maximum performance and efficiency for our users and customers,” the company said. “While not every project moves into production, this rigorous exploration is central to our full stack approach. By co-designing our hardware and software from the ground up, we ensure our systems are integrated and highly optimized for real-world workloads.”

The move fits into a broader pattern taking shape across the AI industry. AI companies have increasingly built their own chips both to run their models more efficiently and to work around global shortages in AI computing capacity. Efficiency has become a key selling point across the industry as concerns about AI spending have cooled the market euphoria that once surrounded it, and firms are trying to wean themselves off Nvidia, which has historically dominated the AI chip market and left major AI makers dependent on its hardware. In June, OpenAI announced its first custom chip, an inference processor called Jalapeño. Earlier this month, it was reported that Anthropic was discussing a new chipmaking partnership with Samsung.

The timing matters for Alphabet specifically. Investors have previously raised concerns about the company’s massive planned AI spending — Google said earlier this year that it plans to spend between $180 billion and $190 billion. With so much money at stake, the company needs to show those investments will pay off. News of the more efficient Frozen v2 chip appears to have reassured investors ahead of Google’s earnings report later this week: following publication of The Information’s report, the company’s stock climbed some 3% on Monday morning.

Why it matters

A markedly more efficient in-house chip would bolster Google’s case that its huge AI infrastructure spending is paying off, and would deepen the broader industry shift away from dependence on Nvidia’s hardware.